International Journal of Computer Applications (0975 – 8887) Volume 126 – No.15, September 2015 13 Frequency and Timing Offset Analysis in OFDM using GNU Radio G. Yamuna M.Tech student ECE Department CVR College of Engineering T. Padmavathi Senior Assistant professor ECE Department CVR College of Engineering Varghese Thattil Prof & Head of ECE Department CVR College of Engineering ABSTRACT To enhance the performance and reduce the Inter Symbol Interference (ISI) at enhanced data rates in wireless communications, Orthogonal Frequency Division Multiplexing (OFDM) is being used. The performance of OFDM is degraded due to frequency and timing offsets which increase the Bit Error Rate (BER) in the wireless communication. Frequency offsets are caused due to difference between transmitter / receiver oscillators and Doppler shift. Timing offsets are due to symbol timing and sampling clock drift. This is estimated and compensated in the receiver. In this paper packet based data transmission test bed has been implemented for OFDM to analyze frequency and time deviation. The effect of frequency offset and timing offset in packet reception over different modulation techniques like BPSK, QPSK is evaluated. By incorporating the noise voltage, normalized frequency and sample timing offset the channel model is simulated in the noisy environment and effect of these parameters considered to maximize the throughput. The OFDM packet based data transmission with offset correction is simulated by using GNU Radio which is free & open-source software tool that provides signal processing blocks with a facility of modifying codes in these blocks. Keywords OFDM, BER, Frequency offset, Timing offset, GNU Radio, SDR. 1. INTRODUCTION The scope of wireless communication is increased to a great extent because of the use of multicarrier modulation techniques, among which Orthogonal Frequency Division Multiplexing (OFDM) is one of the most popular techniques. OFDM is a modulation technique that is being used for many of the latest wireless and telecommunications standards. It allows many users to transmit in an allocated band, by subdividing the available bandwidth into many narrow bandwidth carriers [1]. Each user is allocated several carriers to transmit their data. The transmission is in such a way that the carriers used are orthogonal to one another [2], thus allowing them to be packed together much closer than standard frequency division multiplexing (FDM). This leads to OFDM providing a high spectral efficiency. Additionally the advantage of OFDM is, it reduces fading effect and increases the data rate in transmission [3]. When the signal exists beyond Guard Interval (GI) it leads to Inter Symbol Interference (ISI). Doppler shift in received frequencies destroys the subcarrier’s orthogonality and produces Inter Carrier Interference (ICI) [5] in the system. These ISI and ICI increases the packet losses due to which the performance of an OFDM system to be degraded. The basic model of OFDM communication system with channel model for the transmission of data symbols to the receiver shown in Fig 1. Fig 1: Block diagram of conventional OFDM system This channel model produces LOS (Line of sight) communication and also various reflections due to which multipath effects come to the picture. To minimize the multipath effects and noise introduced by the channel, channel estimation is used. Generally, the received signal obtained through a multipath fading channel contains the channel impulse response and additive white Gaussian noise which is given by, R(n) = H(n)C(n)+W(n) 0 n N-1 (1) Where C(n) is transmitted complex data symbol which is modulated by a BPSK or QAM modulation, H (n) is channel coefficient in frequency domain and W (n) is additive white Gaussian noise. The channel coefficient H (n) is Hn =  h j r−1 j=0 e j2πTf Dj n N  (λ−τ j ) (2) In equation (1.2), λ is the delay spread index, h j is the complex impulse response of the j th path, f Dj is the j th path’s Doppler frequency shift, τ j is the j th path delay time normalized by the sampling time, and r is the total number of the propagation path. After removing the guard interval from R(n) the received samples are obtained and are demodulated to extract the pilot signals. The transmitted data samples C(n)